NVIDIA GeForce RTX 4070 Super
Expert rating8.5 / 10
- GPU Brand:
- NVIDIA
- Architecture:
- Ada Lovelace (AD104)
- Tier:
- Mid-Range
- CUDA Cores:
- 7168
- VRAM:
- 12 GB
- Memory Type:
- GDDR6X
- Ray Tracing Support:
- Yes
- TDP:
- 220 W
Consumercarts is reader-supported. Learn more
Disclaimer: The price and specification of the given product may differ from the actual product. Please confirm on the retailer's website before buying.
Manufacturer: NVIDIA Last updated: September 11, 2026
NVIDIA GeForce RTX 4070 Super (Ada Lovelace/AD104/7168 CUDA Cores/12GB GDDR6X/220W TDP/DLSS 3) Review
The RTX 4070 Super refreshed NVIDIA's mid-range in early 2024 with a meaningfully larger die than the original RTX 4070, closing much of the gap to the RTX 4070 Ti at the same $599 launch price as the card it replaced.
7168 CUDA cores, up from 5888 on the original 4070, deliver a genuine double-digit percentage gain in most games rather than a rebrand, built on the same 12GB/192-bit memory configuration as its predecessor.
A Real Mid-Generation Upgrade
7168 CUDA cores (up from 5888 on the original 4070) deliver a genuine double-digit percentage gain in most games, not just a rebrand. It's built on the same 12GB/192-bit memory configuration as its predecessor.
1440p High-Refresh Sweet Spot
This is the card NVIDIA positions for smooth 1440p at high settings and high refresh rates, with enough headroom for light 4K work using DLSS upscaling.
Comparison to Alternatives
Against the AMD Radeon RX 7800 XT, the 4070 Super usually trades blows in rasterization but pulls ahead in ray tracing and DLSS 3 support. Against the pricier RTX 4070 Ti Super, the standard Super gives up VRAM (12GB vs 16GB) and memory bus width for a lower price.
Specifications
Platform
| GPU Brand | NVIDIA |
|---|
Performance
| VRAM | 12 GB |
|---|---|
| Memory Type | GDDR6X |
| Boost Clock | 2475 MHz |
| Ray Tracing Support | Yes |
Physical & Power
| TDP | 220 W |
|---|
Silicon
| Architecture | Ada Lovelace (AD104) |
|---|---|
| Shader Units | 7168 |
Memory
| Memory Bus Width | 192 bit |
|---|
Market and workstation
| MSRP | 599 USD |
|---|
| Brand | NVIDIA |
|---|---|
| Model | GeForce RTX 4070 Super |
Expert Review
Reviewed by Burooj Alam. Rated by us, not by users.
Expert rating8.5 / 10
A genuine mid-generation upgrade rather than a rebrand, and the Ada generation's most balanced 1440p high-refresh card.
Best for
- 1440p high-refresh gaming with DLSS 3 headroom.
Not for
- Buyers targeting native 4K, where the 12GB buffer can become a bottleneck.
FAQs
What are the key specifications of the NVIDIA GeForce RTX 4070 Super (Ada Lovelace/AD104/7168 CUDA Cores/12GB GDDR6X/220W TDP/DLSS 3)?
The NVIDIA GeForce RTX 4070 Super (Ada Lovelace/AD104/7168 CUDA Cores/12GB GDDR6X/220W TDP/DLSS 3) is a graphics card from NVIDIA. Key specs: GPU Brand NVIDIA, Architecture Ada Lovelace (AD104), CUDA Cores 7168, Boost Clock 2475, VRAM 12 GB.
Who is the NVIDIA GeForce RTX 4070 Super (Ada Lovelace/AD104/7168 CUDA Cores/12GB GDDR6X/220W TDP/DLSS 3) best suited to?
It suits 1440p high-refresh gaming with dlss 3 headroom..
What are the drawbacks of the NVIDIA GeForce RTX 4070 Super (Ada Lovelace/AD104/7168 CUDA Cores/12GB GDDR6X/220W TDP/DLSS 3)?
It is a poor fit for buyers targeting native 4k, where the 12gb buffer can become a bottleneck..
Compare and read more
- AMD Instinct MI210 vs NVIDIA GeForce RTX 4070 Super (Ada Lovelace/AD104/7168 CUDA Cores/12GB GDDR6X/220W TDP/DLSS 3)
- Acer Predator BiFrost OC Arc A770 vs NVIDIA GeForce RTX 4070 Super (Ada Lovelace/AD104/7168 CUDA Cores/12GB GDDR6X/220W TDP/DLSS 3)
- AMD Instinct MI100 vs NVIDIA GeForce RTX 4070 Super (Ada Lovelace/AD104/7168 CUDA Cores/12GB GDDR6X/220W TDP/DLSS 3)
- AMD Instinct MI250 vs NVIDIA GeForce RTX 4070 Super (Ada Lovelace/AD104/7168 CUDA Cores/12GB GDDR6X/220W TDP/DLSS 3)